# Research

We focus on the development of measurement methods and technologies for biomedical and also technical applications of nuclear magnetic resonance and to the application of MR techniques in preclinical, particularly translational research utilizing mouse and rat animal models of neurologic, psychiatric and oncologic diseases. The current research emphasizes:

• Measurement of organ perfusion and pharmacokinetic modeling
• Measurement of water diffusion parameters,
• Spatially localized MR spectroscopy (MRS) and spectroscopic imaging (MRSI), including metabolite quantitation and computer simulation, characterization and tracing of contrast agents in vitro and in vivo, aiming at targeted transport of nanoparticles
• MR-based morphometry and
• Multiparametric data analysis

Ensuring verifiable reproducible quantitative results for specific applications is one of the crucial goals. Other areas of interest include testing MR compatibility of materials and utilization of MR for the study of plants and porous materials. The experimental research is based on a modern 9.4T MR system Bruker Biospec 94/30, equipped for multinuclear preclinical imaging, and an accredited mouse and rat animal facility. Also a 4.7T MR imaging system is available. The group closely collaborates with other subjects (Masaryk University, Veterinary Research Institute and others), which ensures the multidisciplinary needs of preclinical research employing MR. Translation of the techniques to clinical practice is another part of the research interests.

## Perfusion

Perfusion is the physiologic process in which blood is carried by capillaries (diameter 5-15 $\mu$m) to close vicinity (~15 $\mu$m) of cells, providing them, among other, with water, gases, nutrients, hormones, waste removal, immune system services or thermoregulation. Therefore, it is essential for animal cell homeostasis. Hypoperfusion is associated with conditions such as artery diseases, low blood pressure, heart failure, loss of blood volume or tumor necrosis; changed characteristics may be indicative of blood-brain barrier disruption or tumor neoangio-genesis or necrosis. Perfusion is also needed for the delivery of medical drugs to diseased cells. For these reasons, robust quantitative assessment of local perfusion parameters can support pathophysiology research, drug development, diagnostics and therapy planning and monitoring, which we observe in the external demand for this type of measurements. For the physiology-dictated temporal resolution of ~1 s, the achievable resolution is ~300 $\mu$m, which is too coarse to show individual capillaries. However, the aggregate signal dynamics can provide insight in the perfusion processes if suitable experimental and data modeling techniques – including NMR and pharmacokinetics (PK) – are applied to isolate parameters such as blood flow, mean transit time, intravascular plasma and interstitial volumes, and vascular permeability.

The aim of our research work has been to improve the physical specificity, robustness, accuracy and precision of the perfusion characteristics obtained from MR imaging, with emphasis on relevant biomarkers of cancer and stroke.

## Spectroscopy

In vivo 1H MR spectroscopic techniques can distinguish the signals of water, fat, and up to about 20 low-molecular-weight metabolites and to quantify their concen-trations. Such techniques are declared as available on most human and animal MR scanners, but in fact are only marginally supported by the manufacturers, despite the opportunity to detect important metabolic changes, particularly in the brain and tumors. Spectroscopy of other nuclei, particularly 31P and 13C, is also available on some scanners. While 31P offers the possibility to observe energetically important metabolites, observation of 13C-labeled compounds can help in the exploration of more complex metabolic processes. Unfortunately, the low concentrations of metabolite molecules, the presence of strong nuisance signals and interference of spatial and spectral encoding result in low SNR and signals often marred by artifacts, which often complicates inference of sufficiently accurate and precise molecule concentrations. Nevertheless, examples showing the achievability of high-quality MR spectroscopic and even spectroscopic imaging data in human high-field scanners exist.

The aim of our research work has been to analyze the pitfalls and support the global spectroscopic community by contributing to collaborative development of software for quantitative analysis of spectroscopic signals, and by development towards fast spectroscopic imaging in humans and animals.

## Diffusion

MR is the only modality that can noninvasively characterize the local mobility of water molecules in vivo. In 2015-2019, our approach was driven by biomedical research looking for objective biomarkers of Parkinson’s disease. This disease is associated with overexpression of protein $\alpha$-synuclein, which aggregates in some neurons and forms Lewy bodies, which were hypothesized to reduce the cytosolic water diffusion. We indeed observed that in GM-model animals in affected brain areas restricted diffusion could be detected by diffusion kurtosis imaging, DKI, measured with pulse sequences adapted to our instrument conditions and measurement time limit of 3 hours. Data were evaluated by 3rd party software. ÚPT provided the measurements and data evaluation, interpretation was the responsibility of MUNI.

## Direct myelin imaging

We explored the possibility of direct imaging of myelin, which plays an important role in some neurodegenerative diseases. In collaboration with MUNI we developed for a human 3T scanner an imaging technique with extremely short echo-time of 60 $\mu$s, allowing us to detect the fast relaxing signals of myelin. The achievable SNR was impractically low, but this technique is now useful for imaging tendons.

## Nanoparticle tests

We tested in vitro and in vivo experimental contrast agents and potential drug carriers based on graphene as a subcontract of UPOL research.

## Diffusion-reaction observation

In a test tube, we observed the othewise invisible dynamics of a diffusion-reaction process simulating the geologically interesting formation of visible Liesegang patterns.